Page 88 - Copper and Bronze in Art: Corrosion, Colorants, Getty Museum Conservation, By David Scott

the amount of dissolved carbonate, which is strongly influenced by the marine organisms living
in the vicinity of the site.
MacLeod (1991) has noted evidence for the presence of cupric hydroxide, Cu (OH) 2 , in con­
creted marine copper alloys that were freshly excavated. This compound will usually transform
to other products once the object is excavated and allowed to dry because in the presence of
chlorides or carbonate anions a reaction will ensue to produce atacamite or malachite. Brass
objects are often not extensively corroded, despite the popular assumption that widespread
dezincification should be occurring. Perhaps the presence of alloying elements —such as lead,
arsenic, or antimony—may inhibit the dezincification process, as suggested by Lucey (1972).
For bronze alloys, Campbell and Mills (1977) showed that in deaerated seawater, the eutectoid
(alpha+delta) phase was preferentially attacked in a 77Cu20Sn3Pb alloy, leaving the surface
enriched in an alpha phase. At higher oxygen levels, the copper-rich alpha phase was preferen­
tially attacked instead, preserving the alpha+delta eutectoid. MacLeod (1991) confirmed this
general conclusion, based on an excavated bronze bell from the I881 wreck of the Rapid. Part of
the bell's lip and crown was buried in sand and was corroded, with formation of cassiterite,
Sn0 2 , from the tin content of the alloy. The principal copper corrosion product was found to be
paratacamite, with the corrosion front extending 7 mm into the metal. The corrosion on surfaces
of the same object exposed to the sea, however, was only 1 mm thick, and no destannification
had occurred.
Highly leaded bronzes and brasses from these shipwreck sites often show very good corro­
i
sion resistance f they were not buried by sediment. McNeil and Little (1992) discuss aspects of
the corrosion of copper and silver in near-surface environments and note that the solid-state
transformation of cuprite to malachite is possible, while transformation to a copper hydroxy-
chloride requires dissolution and precipitation. The significance of this observation requires
further research. The corrosion rate of copper-nickel alloys in seawater is much slower than the
corrosion of pure copper. The most recent explanation for this is the alteration of the cathodic
reduction potential, which is depressed in the copper-nickel alloys.
I USE OF TREATMENT SOLUTIONS FOR WOOD-METAL ARTIFACTS
Selwyn, Rennie-Bissaillion, and Binnie (1993) investigated the corrosion rate of copper alloys
associated with waterlogged conditions, particularly those alloys in composite wood-metal arti­
facts that had been impregnated with polyethylene glycol 00 (PEG 400) or treated with a range
4
of other consolidants before burial in marine or waterlogged burial sites. These PEG solutions
are mildly acidic, with a pH between 4.5 and 7, and they are capable of causing copper cor­
rosion. In addition, wet wood may corrode contiguous metals because water-soluble wood
extracts, such as formic and acetic acids, produce corrosion, especially of copper and lead alloys.
Selwyn, Rennie-Bissaillion, and Binnie (1993) examined the use of a consolidant, Witcamine
RAD 110o, based on an oxyalkylated rosin amine in a 15 % volume per volume (v/v) solution as
2 0
an alternative treatment for copper-wood composites. Also studied was the effectiveness of

C O R R O S I O N AN D E N V I R O N M E N T
71